Quick & dirty temperature sensing
amxhobbyist
Posts: 90
in AMX Hardware
If I wanted to put some temperature sensors on an Axcess master and get ~1/4 degree (or better) increment readings, what's the easiest way to do this? I assume a sensor on a voltage input board, but does anyone have recommendations for specific sensors and/or code examples?
I'm considering dropping the idea of using communicating thermostats throughout my coach, instead putting the logic in code, controlling the equipment with relays and adjusting setpoints via touchpanel. Then I could do other neat things the thermostats can't, like provide "boost heating" that activates multiple sources of heat at once, auto-switch between heat sources based on any logic I like, etc.
This would all be very easy but I lack the experience in temperature sensing, so any advice would be wonderful. (Side note: If it's easier to do in Netlinx, I'll buy a Netlinx master for the coach - I just happen to have a pile of Axcess stuff laying around from upgrading in the house and was going to re-use it).
I'm considering dropping the idea of using communicating thermostats throughout my coach, instead putting the logic in code, controlling the equipment with relays and adjusting setpoints via touchpanel. Then I could do other neat things the thermostats can't, like provide "boost heating" that activates multiple sources of heat at once, auto-switch between heat sources based on any logic I like, etc.
This would all be very easy but I lack the experience in temperature sensing, so any advice would be wonderful. (Side note: If it's easier to do in Netlinx, I'll buy a Netlinx master for the coach - I just happen to have a pile of Axcess stuff laying around from upgrading in the house and was going to re-use it).
0
Comments
I doubt that you could get the accuracy you are looking for using analog voltages.
I would look for some sort of communicating sensor that you could poll digitally (serially) for the current temperature. /+.25 degrees is quite a small tolerance.
Check Automation Direct. They have quite a selection of such things.
We had a similar issue controlling 6 circuits of pressure lamps in a tunnel and the light sensors suffered the same problem as you have with temp. We ended up sensing with 4 sensors and doing some involved calcs using data from all four, then implementing a hysteresis process to stop them trying to switch levels all the time as the level crossed the trigger point.
A few lux to the naked eye makes no difference, but if that level change is across a trigger point in code, the switching would drive the pressure lamps (and their cool down cycles) mad!
Netlinx processor won't make any difference - the issue as I see it is the input device accuracy, not the ability to perform the calculations.
I am not familiar with any smart sensors, but they would definitely be the go.
What I don't know (or perhaps don't understand) about your situation is the range over which you are trying to sample.
If you are trying to get accuracy over a few degrees, then 8 bit is probably fine. If it is over 30 degrees or more, then my previous post stands!
Yeah - I'd like full accuracy over any temps the coach could reach. So, 0 to 100 degrees in 1/4 degree increments. Actually, 1 degree increments would probably work, but I like accuracy.
I'll look for a sensor that can communicate via rs232, then. If anyone has any leads on such a thing, please let me know. In the meantime I'll find this Automation Direct place and see what they have.
Thanks for the responses!
http://store.qkits.com/moreinfo.cfm/VK011
Used it several times with no issues. Up to 4 sensors, standard ASCII feedback over RS232. Short response showing temp, or long response showing temp, high temp, & low temp. Adjustable poll time from 1 second to 99 minutes. F or C degrees..
And it's only $40 US.
Kevin D.